Satellites in geostationary orbits (GSO's) have been widely preferred because of the economic advantages afforded by such orbits. In a geostationary orbit, a satellite traveling above the earth's equator, in the same direction as that in which the earth is rotating, and at the same angular velocity, appears stationary relative to a point on the earth. These satellites are always “in view” at all locations within their service areas, so their utilization efficiency is effectively one-hundred percent. Antennas on earth need be aimed at a GSO satellite only once; no tracking system is required.
Coordination between GSOs and with terrestrial services is facilitated by governmental allocation of designated “slots” angularly spaced according to service type.
Given the desirability of geostationary satellite orbits and the fact that there are only a finite number of available “slots” in the geostationary “belt,” the latter capacity has been essentially saturated with satellites operating in desirable frequency bands up through the Ku-band (up to 18 GHz). As a result, the government has been auctioning the increasingly scarce remaining slots.
The bottleneck in ground-to-satellite communications may be overcome by increasing the number of RF beams on board of a single satellite, increasing the number of satellites, e.g. deploying those using low earth orbits (LEO's), medium earth orbits (MEO's), or by putting several satellites into a single GSO slot and by using higher frequencies, for example, the Ka band (up to approximately 40 GHz). This appears to be a limit on the number of RF antennas on board of a single satellite. At this point, 50–100 antennas. Growth to higher frequencies is limited by difficult problems in technology and propagation. Expansion in satellite applications requires exploitation of the spatial dimension (i.e., above and below the GSO belt). A host of proposed LEO and MEO systems exemplify this direction.
Therefore, the only remaining way for increasing the capacity of satellite communication systems is increasing the number of the satellites. In this approach, the satellites are interconnected into a network that serves a wide geographic area. Today, laser communication links are planned for intersatellite communications. The advantage of optical intersatellite links over RF links derives from (i) reduced power consumption and (ii) considerably smaller size and weight of an optical telescope versus an RF antenna. As a result, a single satellite can house more communication links, thereby increasing the overall data-handling capacity.
Satellite communications systems employing multiple RF ground links and optical intersatellite links will use complicated switching electronics to route the ever increasing volumes of data traffic. Systems that are being developed include a router that acts as a high speed switch. All data whether optical or RF uplink or downlink signals are converted to the electrical domain and routed appropriately through the satellite. The high speed switching electronics are enlarged to accommodate the optical signals.
High speed switching electronics consume a significant amount of electric power which is always at a premium on board of satellites. Moreover, as the volume of data traffic transmitted by numerous RF and optical channels of a satellite increases, the electronic switch becomes the bottleneck that limits the overall network capacity.
Passive optical routing (which retains and redirects signals destined for further relaying to other satellites in optical domain without down conversion to electronic format) is an attractive way of unloading the on-board electronic switch. Such passive optical routing requires designated optical carriers for each pair of communicating satellites on any intersatellite link. The benefits of passive optical routing include: i) increased network handling capacity due to unloading the electronic switch, ii) transparency to communication protocols, i.e., the intermediate satellites do not have to understand the nature of the signal in order to route it, iii) non-blocking connectivity, i.e., a data stream is not required to wait until a communication link is done with transmitting data to another satellite.
Unfortunately, optical routing may be limited by the number of optical carriers available in a network. As networks become large, a significant amount of optical carriers would be used if only one optical carrier frequency could be used for only one interconnection in the system. It would therefore be desirable to reduce the number of optical carriers by using a unique optical carrier for each overlapping network path. This allows non-overlapping paths to reuse the same optical carrier.